In today's digital era, various real-world applications generate data in streams, and these data streams are of two types stationary data streams, which are static, and non-stationary data streams that are dynamic...
详细信息
This research presents a new key information extraction algorithm from shopping receipts. Specifically, we train semantic, visual and structural features through three deep learning methods, respectively, and formulat...
详细信息
Monitoring electric vehicles’ battery situation and indicating the state of health is still challenging. Temperature is one of the critical factors determining battery degradation over time. We have collected more th...
详细信息
Toddler activity recognition aims in finding out the activities of toddler and also toddlers based on the video captured by camera. The purpose of activity recognition is to monitor a new employee of the company (mayb...
详细信息
Toddler activity recognition aims in finding out the activities of toddler and also toddlers based on the video captured by camera. The purpose of activity recognition is to monitor a new employee of the company (maybe a software or even a restaurant) whether employees are working properly or do we need to give more training. This system can also be used by the parents to monitor the activity of the kids who are alone at home. This system can also be implemented in hospitals to monitor the patients and to check their improvement. Toddler activity recognition system is developed using CNN and LSTM. To predict the activity the path of the video file is given as input. The predicted result by the model is checked for accuracy. To make the result available for the end user (who needs to monitor the activity) and can access the result through two ways. In the First method user can directly view on the system where the model is predicting the result. In the second method user is out of station, so the predicted result will be updated on the cloud on fixed time intervals and user can view the result from the working place itself.
In recent years, the concept of smart spaces and occupancy detection is crucial for creating more efficient and sustainable buildings that prioritize both energy efficiency and occupant comfort. By leveraging advanced...
详细信息
Manufacture and sale of counterfeit goods has been plaguing several business sectors for many years. The production of counterfeit goods are increasing, this has an impact on the sales and earnings of the businesses. ...
Manufacture and sale of counterfeit goods has been plaguing several business sectors for many years. The production of counterfeit goods are increasing, this has an impact on the sales and earnings of the businesses. By stifling revenues, economic growth, and consumer health, it has an impact on governments, global trade, businesses, and consumers. The current technology used for combating counterfeit goods rely on a centralised authority. Issues with this design include single point processing, failure, and storage. Blockchain technology has can be used as an alternative to address all these issues for these *** code can be used along with a blockchain hash code to uniquely identify every single product that ever enters the supply chain, hence no fake product can flood the supply chain. Ethereum blockchain is utilized and smart contracts are deployed to store the data about each and every product. Each products whole history can be tracked by using its QR code.
This research explores the submission of deep neural networks (DNNs) to enhance corresponding product references in e-commerce stages. By leveraging the Neural Collaborative Filtering (NeuMF) model, which integrates C...
详细信息
ISBN:
(数字)9798331518592
ISBN:
(纸本)9798331518608
This research explores the submission of deep neural networks (DNNs) to enhance corresponding product references in e-commerce stages. By leveraging the Neural Collaborative Filtering (NeuMF) model, which integrates Comprehensive Matrix Factorization (GMF) and Multi-Layer Perceptron (MLP), the study captures both linear and non-linear user-item connections to improve recommendation accuracy. Key performance metrics, including accuracy, loss, ROC curve, and precision-recall curve, were analyzed to assess the model's effectiveness. Results show that the NeuMF model significantly improves recommendation accuracy and user satisfaction, with an AUC of 0.94. However, the precision-recall analysis highlights areas for improvement, mainly in handling imbalanced datasets. This work demonstrates the potential of advanced neural models to drive better user engagement and increase sales in dynamic e-commerce environments while suggesting future directions for optimizing precision-recall trade-offs.
Home Automation system with web server deployment is developed using Raspberry Pi Pico and micropython, which turns ON the lights based on the count of persons inside that particular room. The system uses Infrared Red...
Home Automation system with web server deployment is developed using Raspberry Pi Pico and micropython, which turns ON the lights based on the count of persons inside that particular room. The system uses Infrared Red (IR) sensor, which detects the heat radiation in the environment. This heat radiation is not constant as it changes due to the movement of people in the sensing environment. This change is used as a measure to detect the changes in the environment which allows us to perform a particular action for the event. The data is sent to node which sends the data to the webserver running in the Raspberry pi pico. Based on the result, the light is turned ON or OFF.
In this work, we propose a small-size and low-cost phase shifter based on defective microstrip structure (DMS) technique, with a modified reconfigurable unit cell (MRDMS) for WLAN applications at 5.2 GHz. The phase sh...
详细信息
The future of home appliances in the IoT (Internet of Things) lies in self-awareness, independent condition-action rule implementation, and device autonomy. In the concept of autonomy, agent-based IoT appliances can S...
详细信息
ISBN:
(数字)9798331529222
ISBN:
(纸本)9798331529239
The future of home appliances in the IoT (Internet of Things) lies in self-awareness, independent condition-action rule implementation, and device autonomy. In the concept of autonomy, agent-based IoT appliances can SLEEP and WAKE-UP independently as when necessary, perceive and compute data; and engage in communication or negotiation with other agent-based devices. In the paradigm of OOP, this paper describes various home appliances' autonomous artificial intelligent (AAI) behavior. While self-awareness borders on the appliance knowing its current status and location, self-control relates to the appliance's ability to control its outgoing and incoming data and take the appropriate action based on some given environment data conditional rules. Thus, this paper presents some IoT home appliances' operational condition-action behavior (CAB). Firstly, by simulation, then followed by the actual transformation of the CAB rules into modular programs in object-oriented paradigm for sensor and appliance agents. The WindSpeedSensor, TempSensor, and HumiditySensor utilize the given environment data input to communicate with their respective WindowAppliance and CoffeeMachineAppliance agent, as demonstrated in this paper. In turn, the appliance agents take the decision reached by their sensors to perform some autonomous action. This model depicts independent AAI behavior of home appliances for the benefit of home owners.
暂无评论